You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@druid.apache.org by GitBox <gi...@apache.org> on 2020/08/18 15:46:47 UTC

[GitHub] [druid] gianm commented on issue #10042: com.yahoo.sketches.SketchesArgumentException: Incompatible Seed Hashes. 0, -27700

gianm commented on issue #10042:
URL: https://github.com/apache/druid/issues/10042#issuecomment-675558548


   The trace above indicates the Broker is trying to deserialize a sketch sent to it by a data server (like a Historical) as the result of a partial aggregation. So, it's not directly from something stored in a segment. It's been run through an aggregator.
   
   @AlexanderSaydakov, @leerho, is it possible that there is some issue when a sketch written with DataSketches 1.1.0 (as the upgraded Historical would do) is read by DataSketches 0.13.3 (as the Broker would do)?
   
   @capistrant, did you get the same stack trace as @DockerLive? If not could you post yours too?
   
   Btw, you should be able to get @leerho the image he asked for by sending the Historical the same query that the Broker did. You'll get a base64 encoded version of the sketch. The best way to get the same query the Broker issued is probably to either:
   
   - use wireshark to capture it and then reissue it
   - enable request logging on one of your Historicals and then copy the query out of the request log


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org